Skip to main content

You are here

Volatility reduction: How minimum variance indexes work

Minimum variance indexes, which apply rules-based methodologies with the aim of minimizing an index’s volatility, are popular among market participants interested in smart beta. In this FTSE Russell Insights, the third in a series of three covering risk-based indexes and their applications, we review minimum variance indexes in detail.

We discuss the reasons for the recent popularity of the minimum variance approach. We also note the importance of the design choices facing the creator of a minimum variance index, which can determine the index’s suitability for use by market participants as a benchmark or as the underlying target for an index-replicating portfolio or financial product.

Using the example of a minimum variance index constructed from the Russell 1000® Index, we also review the reasons for why a minimum variance strategy has captured reduced volatility in index levels and examine its recent valuation levels.

Why the interest in volatility reduction?

In an environment of near-zero official interest rates and low bond yields, investors increasingly rely on the equity market to meet long-term return targets. At the same time, the bear markets of 2000-2002 and 2007-2008 are a reminder of the downside risks inherent in equity investing.

The FTSE Russell Smart Beta Survey 2016, which gathered responses from more than 250 large asset owners worldwide with over US$2 trillion under management, showed that, after return enhancement, risk reduction was the most popular investment objective cited by investors evaluating smart beta . Smart beta is a generic term for indexes using alternative weightings to benchmark specific risk and return goals.

Some categories of investor, for example European insurance companies subject to the “Solvency II” Directive, are increasingly required to reflect their liability profiles in their asset portfolios, incentivizing them to find solutions and tools  that help manage downside risks while also seeking returns.

Minimum variance indexes apply a rules-based methodology to constrain volatility in a reference index while also maintaining exposure to the equity market. Asset owners have therefore employed these tools as part of their own risk-reduction strategies. At the same time, minimum variance index methodology is not designed to benchmark a specific return objective: any deviation in index values from the capitalization-weighted reference benchmark are an incidental outcome of the volatility-minimizing strategy.

The efficacy of a minimum variance index in reducing volatility can be seen from the chart and table below where we compare the past volatility of the FTSE Developed Minimum Variance Index with the volatility of its capitalization-weighted reference index, the FTSE Developed Index. The FTSE Developed Index measures the performance of mid and large cap stocks in global developed markets and represents around 98% of the world’s investable market capitalization.

Exhibit 1 demonstrates that, over the period between 2001 and 2016, the minimum variance approach consistently had a 20-30 percent index volatility reduction by comparison with the capitalization-weighted reference benchmark.

The minimum variance index also suffered a smaller maximum drawdown in the 2007 - 2008 bear market than the reference benchmark (see Exhibit 2). And it shared to a greater extent in the equity market’s upward moves than in its downward moves (expressed in the table as a higher up capture ratio than down capture ratio).

Competing approaches to volatility reduction

Minimum variance is not the only index construction approach focused on capturing reduced equity market volatility in its index composition.

A simple alternative low volatility approach is a risk-weighted index, where the index’s constituents are weighted by the inverse of their historical volatility (the lower the volatility, the higher the index weighting).

Another approach is to construct a low volatility factor index. Here, the index methodology focuses on capturing the factor premium historically associated with low volatility stocks.

A comparison of the historical index-level volatility reduction resulting from these three approaches, based on the capitalization-weighted Russell 1000 Index universe, shows that minimum variance has produced higher levels of index-level volatility reduction than the two alternative approaches over an extended period (see Exhibit 3).

It’s noteworthy that both the simple risk-weighted (inverse volatility) approach and the low volatility factor approach achieved relatively low levels of index-level risk reduction over the period (typically 0–10%).

The minimum variance approach succeeded in producing higher levels of risk reduction primarily because, as well as focusing on stocks’ past volatilities, it fully exploited the correlation structure of the equity market.

In fact, minimum variance indexes may select individual stocks with higher levels of volatility if this produces a correlation (i.e., a risk reduction) benefit to the overall index—a notable difference with the risk-weighted approach or the low-volatility factor approach.

—Download the paper—